German methane fluxes estimated top-down using ICON–ART – Part 1: Ensemble-enhanced scaling inversion

Abstract. This two-part study explores the quantification of greenhouse gas emissions using atmospheric observations in order to validate national emission inventories. Inverse methods can support emission quantification at the national scale based on observations and atmospheric transport simulations, yet, they are often limited by the observation coverage, transport model uncertainties, and inversion methodologies. Here, we introduce a system for regional estimation of methane fluxes and apply this to Central Europe with a focus on Germany, where we distinguish emissions from different anthropogenic sectors. We evaluate the robustness of the method using sensitivity tests with in-situ observations from the Integrated Carbon Observation System (ICOS). Using synthetic observation experiments, we estimate the impact of transport errors on the flux estimates. The atmospheric transport is calculated employing the numerical weather prediction model ICON with its module ART at 6.5 km resolution, sampling the meteorological uncertainty with a 12-member transport ensemble. The same transport ensemble is used to generate pseudo-observations with a simulated transport uncertainty. Posterior fluxes are estimated with a synthesis inversion method for three different approximations of the model–observation error covariance matrix. We find that using ensemble-estimated transport uncertainties can significantly reduce the random error of emission estimates. Our results highlight the importance of analyzing biases in flux inversions for reliable, observation-based emission estimates.